CTCarmen D. Tekwe

Carmen D. Tekwe

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Carmen D. Tekwe has a total of 153 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Reducing bias and enhancing equity in AI-enabled precision nutrition: addressing measurement error across wearables, multiomics, and dietary data

    by Andi Mai, Yuanyuan Luan, See Ling Loy, Caihong Qin, Heyang Ji, Ashley Obeng, Mercy Oladuti, Keisuke Ejima, Kai Ting Mok, Satvinder Kaur, et al.

    Artificial intelligence (AI) can offer individualized dietary guidance based on multimodal data collected from various sources, including wearable sensors, high-dimensional multiomics and biomarker analyses, behavioral tracking, and self-reported dietary intake, enabling the emergence of precision nutrition. However, the predictive power and fairness of these models rely on the quality of the data inputs, and measurement errors in any of these underlying data streams can introduce systematic bias, degrade model performance, and disproportionately affect underserved populations. In this review, we examine the central role played by measurement error in AI-driven nutrition tools and evaluate statistical and machine learning approaches for mitigating the impacts of measurement error. We provide structured comparisons exploring both classical methods (e.g., regression calibration, Bayesian models) and emerging AI strategies (e.g., denoising autoencoders, multitask learning, uncertainty-aware deep learning) for correcting biased inputs. We also explore how uncorrected measurement error can perpetuate demographic biases, compromise efforts toward personalized medicine, and exacerbate equity gaps when models are deployed in real-world settings. Our review draws upon evidence across nutrition science, digital health, and algorithmic fairness. We propose a framework and offer actionable strategies for overcoming measurement error that can be implemented by researchers, developers, and regulators working at the intersection of data science and dietary health and seeking to build calibration-aware, inclusive precision nutrition systems.


  • Correcting Measurement Error and Zero Inflation in Functional Covariates for Scalar‐on‐Function Quantile Regression

    by Caihong Qin, Lan Xue, Ufuk Beyaztas, Roger S. Zoh, Mark Benden, Jeff Goldsmith, Carmen D. Tekwe

    Wearable devices collect time‐varying biobehavioral data, offering opportunities to investigate how behaviors influence health outcomes. However, these data often contain measurement error and excess zeros (due to nonwear, sedentary behavior, or connectivity issues), each characterized by subject‐specific distributions. Current statistical methods fail to address these issues simultaneously. We introduce a novel modeling framework for zero‐inflated and error‐prone functional data by incorporating a subject‐specific time‐varying validity indicator that explicitly distinguishes structural zeros from intrinsic values. We iteratively estimate the latent functional covariates and zero‐inflation probabilities via maximum likelihood, using basis expansions and linear mixed models to adjust for measurement error. To assess the effects of the recovered latent covariates, we apply joint quantile regression across multiple quantile levels. Through extensive simulations, we demonstrate that our approach significantly improves estimation accuracy over methods that only address measurement error, and joint estimation yields substantial improvements compared with fitting separate quantile regressions. Applied to a childhood obesity study, our approach effectively corrects for zero inflation and measurement error in step counts, yielding results that closely align with energy expenditure and supporting their use as a proxy for physical activity.

  • Generalized Functional Linear Regression Models With Functional and Scalar Covariates Prone to Measurement Error

    by Yuanyuan Luan, Roger S. Zoh, Sneha Jadhav, Lan Xue, Carmen D. Tekwe

    Most methods for adjusting for biases due to measurement errors in covariates in generalized linear regression models focus on scalar covariates. Less work exists to correct for biases due to measurement error in a mixture of functional and scalar covariates. To address this limitation, we develop joint functional simulation extrapolation (FSIMEX) and mixed effects model‐based (MEM) approaches in a mixture of functional and scalar covariates prone to classical measurement errors in generalized functional linear regression. The proposed methods were compared with the Oracle estimator, which served as a benchmark, principal components analysis through the conditional expectation (PACE), Naive_ave, and Naive_one methods through extensive simulations to assess their finite sample performance. Across a range of conditions, the joint FSIMEX estimator generally had low biases, followed by the MEM estimator, similar to those of the Oracle estimator. The SIMEX and MEM methods for measurement error adjustment performed notably better than the Naive_one estimators that did not adjust for measurement error. Although the Naive_ave and PACE estimators adjusted for some of the measurement errors, they did not account for the heteroscedastic covariance structures associated with the functional covariate and the PACE approach was not designed for modeling scalar predictors. We applied our methods to data from the 2011–2014 cycles of the National Health and Examination Survey (NHANES) to assess the relationships physical activity, total caloric intake and demographic characteristics have with type 2 diabetes status of community dwelling adults living in the United States. We treated the device‐based measure of physical activity as a functional covariate prone to complex arbitrary heteroscedastic errors and total caloric intake as a scalar covariate prone to error. Estimation with the joint FSIMEX and MEM estimators reduced the associations for some covariates but increased them for others, particularly physical activity, in comparison to estimation with the estimators that did not adjust for biases due to measurement error. Overall, we observed that failing to account for measurement error due to biases can lead to biased estimations of the functional and scalar covariates, which are prone to errors in the generalized functional linear regression models.

  • Open Access

    Least squares-based methods to bias adjustment in scalar-on-function regression model using a functional instrumental variable

    by Xiwei Chen, Ufuk Beyaztas, Caihong Qin, Heyang Ji, Gilson Honvoh, Roger S. Zoh, Lan Xue, Carmen D. Tekwe

    Instrumental variables are widely used to adjust for measurement error bias when assessing associations of health outcomes with ME prone independent variables. IV approaches addressing ME in longitudinal models are well established, but few methods exist for functional regression. We develop two methods to adjust for ME bias in scalar on function linear models. We regress a scalar outcome on an ME prone functional variable using a functional IV for model identification and propose two least squares based methods to adjust for ME bias. Our methods alleviate potential computational challenges encountered when applying classical regression calibration methods for bias adjustment in high dimensional settings and adjust for potential serial correlations across time. Simulations demonstrate faster run times, lower bias, and lower AIMSE for the proposed methods when compared to existing approaches. The proposed methods were applied to investigate the association between body mass index and wearable device-based physical activity intensity among community dwelling adults living in the United States.

  • Spatial Function-on-Function Regression

    by Ufuk Beyaztas, Han Lin Shang, Gizel Bakicierler Sezer, Abhijit Mandal, Roger S. Zoh, Carmen D. Tekwe


  • Comparison of two measurement error–correction approaches for assessing the association between choline intake and coronary heart disease prevalence among US community-dwelling adults

    by Cuiqiong Huo, Yuanyuan Luan, Roger S. Zoh, Nana Gletsu-Miller, Stephen J. Carter, Georgia Frey, Hsien-Chang Lin, Aurelian Bidulescu, See Ling Loy, Marwah Abdalla, et al.


  • Open Access

    Functional Linear Cox Regression Model with Frailty

    by Deniz Inan, Ufuk Beyaztas, Carmen D. Tekwe, Xiwei Chen, Roger S. Zoh

    This paper presents a functional linear Cox regression model with frailty to tackle unobserved heterogeneity in survival data with functional covariates. While traditional Cox models are common, they struggle to incorporate frailty effects that represent individual differences not captured by observed covariates. Our model combines scalar and functional covariates with a frailty term to address these unmeasured influences, creating a robust framework for high-dimensional survival analysis. We estimate parameters using functional principal component analysis and apply penalized partial likelihood for the frailty structure. A simulation study shows that our model outperforms traditional approaches in estimation accuracy and predictive capacity, especially with high frailty. We also analyze data from the National Health and Nutrition Examination Survey, highlighting significant links between physical activity and mortality in frail subpopulations. Our findings demonstrate the model's effectiveness in managing complex survival data, with potential applications in biomedical research related to unobserved heterogeneity. The method is available as an R package.

  • Open Access

    Spatial function-on-function regression

    by Ufuk Beyaztas, Han Lin Shang, Gizel Bakicierler Sezer, Abhijit Mandal, Roger S. Zoh, Carmen D. Tekwe

    We introduce a spatial function-on-function regression model to capture spatial dependencies in functional data by integrating spatial autoregressive techniques with functional principal component analysis. The proposed model addresses a critical gap in functional regression by enabling the analysis of functional responses influenced by spatially correlated functional predictors, a common scenario in fields such as environmental sciences, epidemiology, and socio-economic studies. The model employs a spatial functional principal component decomposition on the response and a classical functional principal component decomposition on the predictor, transforming the functional data into a finite-dimensional multivariate spatial autoregressive framework. This transformation allows efficient estimation and robust handling of spatial dependencies through least squares methods. In a series of extensive simulations, the proposed model consistently demonstrated superior performance in estimating both spatial autocorrelation and regression coefficient functions compared to some favorably existing traditional approaches, particularly under moderate to strong spatial effects. Application of the proposed model to Brazilian COVID-19 data further underscored its practical utility, revealing critical spatial patterns in confirmed cases and death rates that align with known geographic and social interactions. An R package provides a comprehensive implementation of the proposed estimation method, offering a user-friendly and efficient tool for researchers and practitioners to apply the methodology in real-world scenarios.

  • MECfda: Scalar-on-Function Regression with Measurement Error Correction

    by Heyang Ji, Ufuk Beyaztas, Yuanyuan Luan, Xiwei Chen, Mengli Zhang, Roger Zoh, Lan Xue, Carmen Tekwe